Enhancing the U.S.-China AI Discourse: Incorporating Multiple Participants for Optimal Outcomes
A New Era of Collaboration: Harnessing the Power of Track 1.5 and Track 2 Dialogues in AI Safety
The United States and China have embarked on their inaugural intergovernmental dialogue about artificial intelligence (AI) in Geneva, May 14, 2024. The discussion revolved around shared concerns and domestic approaches to mitigate AI risks and govern issues related to advanced systems. This dialogue marks a significant step towards fostering cooperation between the two nations, but with the ever-growing prevalence of both commercial and military AI applications, more coordination is essential, particularly among nongovernmental stakeholders.
As the world's top economic powerhouses and leading players in AI, the United States and China wield considerable influence in shaping the future of AI. When U.S. President Joe Biden and Chinese President Xi Jinping agreed to commence these AI-focused intergovernmental talks last November, the path towards reaching a mutual understanding of AI risks and governance frameworks was laid. The challenge now lies in transforming this initial agreement into meaningful and sustainable action.
Collaboration faces several obstacles. Persistent geopolitical tensions and mistrust between these two countries may slow international cooperation and lead to ineffective countermeasures concerning AI risk mitigation, jeopardizing global agreements. Furthermore, diplomatic processes often struggle to keep pace with the swift advancement of AI technologies.
Promoting direct collaboration among experts from both nations is vital to bridge understanding gaps and speed up the development of evidence-based policymaking. Setting aside formal "track 1" diplomatic channels, hosting track 1.5 and track 2 dialogues involving non-governmental experts can foster trust, generate innovative solutions, and forge consensus among various nations. Track 1.5 dialogues involve both government officials and non-governmental experts, while track 2 dialogues consist exclusively of unofficial representatives. These conversations occur frequently between stakeholders in China and the United States to maintain stable relations. The advantage of Track 2 diplomacy lies in its flexibility, enabling discussions on cutting-edge and sensitive issues. Participants from the AI sector can explore consensus on addressing risks through global norms and technical measures.
However, an analysis of Track 2 dialogues on AI in recent years has revealed that most participants are foreign policy and military professionals. The involvement of content experts and industry representatives is scarce, resulting in a narrow emphasis on geopolitical and national security concerns, while neglecting AI technology development's technical and broader aspects. Scientists and industry experts hold critical knowledge that is crucial for providing a nuanced understanding of AI technology development and application under real-world scenarios. Involving these stakeholders can shift the conversation from high-level political negotiations to technical problem-solving, increasing the likelihood of generating practical solutions to common challenges.
The inclusion of experts from academia and industry is essential for several reasons. First, scientists and industry leaders are the vanguard of AI research, development, and implementation, possessing the technical expertise required to identify, foresee, and address intricate AI safety issues. They can analyze specific AI deployments and work on resolving technical issues. Second, collaboration among scientists can aid in depoliticizing the dialogue, inspiring conversations centered on shared goals of advancing knowledge and technology about AI safety, such as safety benchmarks and testing protocols. Academic experts tend to focus on technical aspects and contributions instead of discussing existential risk with an alarmist attitude. Third, given the vital roles AI companies play in both China and the United States, they can provide valuable input on technology research and development, technical measures, and governance frameworks. All in all, expanding the dialogue to include a more diverse group of stakeholders can offer various perspectives on the opportunities and challenges associated with AI, potentially fostering innovation and experimentation while ensuring AI applications are safe and advantageous. For instance, they could embark on joint research projects regarding specific AI safety issues, like algorithmic bias and deepfakes, leading to breakthroughs that benefit both nations and the global community.
Though the intergovernmental dialogue between China and the United States on AI safety is an auspicious beginning, the initiation of Track 1 dialogue does not diminish the importance of Track 1.5 and Track 2 dialogues. Conversely, these dialogues can play an even more influential role, stimulating and catalyzing Track 1 dialogue to progress in a comprehensive and in-depth manner, including fostering cooperation between scientists and the private sector. By engaging a wider range of stakeholders and promoting international cooperation, the two countries can pioneer the development of a robust, ethical, and safe AI framework that benefits the global society.
Image credits: Rawpixel*
Leveraging the Advantages of Track 1.5 and Track 2 Dialogues in AI Safety Collaboration
Track 1.5 Dialogues
- Cross-disciplinary Engagement: Track 1.5 dialogues bring together government representatives and non-governmental experts, fostering a deeper understanding and cooperation on AI safety matters. This combination of viewpoints can result in more all-encompassing solutions.
- Increased Transparency: By involving both government and non-governmental experts, these dialogues can foster transparency on national AI policies and strategies, reducing misperceptions and boosting trust.
- Practical Collaboration: They facilitate practical collaborations on AI safety research and governance, enabling the development of collaborative strategies to mitigate AI risks.
Track 2 Dialogues
- Expert-driven Discussions: Track 2 dialogues, led by non-governmental experts, serve as a platform for detailed technical discussions about AI safety challenges, without the formality of government-level meetings.
- Innovative Solutions: These dialogues can lead to innovative solutions by tapping into the expertise of researchers and industry leaders from both countries.
- Building Trust: By fostering personal relationships among experts, Track 2 dialogues help build trust and understanding, which is vital for sustained collaboration on sensitive issues like AI safety.
- To bridge the gap in understanding and hasten the development of practical solutions, collaborations amongst experts from both the United States and China are necessary, particularly in the realm of evidence-based policymaking.
- Integrating non-governmental experts into Track 1.5 dialogues can enhance transparency on national AI policies, reduce misperceptions, and ultimately boost trust.
- Track 2 dialogues, spearheaded by unofficial representatives, provide an ideal platform for in-depth technical discussions about AI safety concerns, which can eventually lead to innovative solutions and the forging of consensus.
- The participation of scientists and industry experts is crucial for solving intricate AI safety issues, depoliticizing the dialogue, and ensuring discussions remain centered on advancing AI technology for the benefit of both nations and the global community.
